US7447810B2 - Implementing bufferless Direct Memory Access (DMA) controllers using split transactions - Google Patents

Implementing bufferless Direct Memory Access (DMA) controllers using split transactions Download PDF

Info

Publication number
US7447810B2
US7447810B2 US10975803 US97580304A US7447810B2 US 7447810 B2 US7447810 B2 US 7447810B2 US 10975803 US10975803 US 10975803 US 97580304 A US97580304 A US 97580304A US 7447810 B2 US7447810 B2 US 7447810B2
Authority
US
Grant status
Grant
Patent type
Prior art keywords
command
data
controller
read
memory
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active, expires
Application number
US10975803
Other versions
US20060095604A1 (en )
Inventor
Samantha J. Edirisooriya
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Intel Corp
Original Assignee
Intel Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Grant date

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F13/00Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
    • G06F13/14Handling requests for interconnection or transfer
    • G06F13/20Handling requests for interconnection or transfer for access to input/output bus
    • G06F13/28Handling requests for interconnection or transfer for access to input/output bus using burst mode transfer, e.g. direct memory access DMA, cycle steal
    • YGENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02DCLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THIR OWN ENERGY USE
    • Y02D10/00Energy efficient computing
    • Y02D10/10Reducing energy consumption at the single machine level, e.g. processors, personal computers, peripherals or power supply
    • Y02D10/14Interconnection, or transfer of information or other signals between, memories, peripherals or central processing units

Abstract

According to one embodiment a method for implementing bufferless DMA controllers using split transaction functionality is presented. One embodiment of the method comprises, generating a write command from a disk controller directed to a destination unit, the write command including an identifier, generating a read command from the disk controller directed to a source unit, the read command including an identifier which matches the identifier in the write command, the source unit transmitting read data on a split transaction bus, the read data including the identifier of the read command, and receiving the read data at the destination unit via the split transaction bus if the identifier of the read data matches the identifier of the write command.

Description

FIELD OF THE INVENTION

The present embodiments of the invention relate generally to input/output (I/O) processors and, more specifically, relate to direct memory access (DMA) controllers.

BACKGROUND

Many storage, networking, and embedded applications require fast input/output (I/O) throughput for optimal performance. I/O processors allow servers, workstations, and storage subsystems to transfer data faster, reduce communication bottlenecks, and improve overall system performance by offloading I/O processing functions from a host central processing unit (CPU).

Typically, the CPU(s) in the I/O processors program direct memory access (DMA) controller(s) to move data between specified sources and destinations, such as between local memory and host memory. Once the DMA controller is programmed, it will generate a read command to the source's interface or controller. This controller or interface will generate the read command for the source, and once it obtains the read data will place that data on the bus to the DMA controller. Typical DMA controllers include buffers to temporarily store data when the data is moved between sources and destinations, such as between host and local memories. The DMA controller will accept the read data and store it in the DMA controller data buffers. At this time, the DMA controller will generate a write command to the destination's interface or controller. The destination interface or controller will accept this write command. Finally, the DMA controller provides the write data being stored in the DMA controller data buffers to the destination interface or controller in order to be written to the destination.

The use of DMA controller data buffers can lead to increased area requirements, increased power requirements, and added complexity to the I/O processor. The use of DMA controller data buffers also slows down performance and increases costs for I/O processors.

BRIEF DESCRIPTION OF THE DRAWINGS

The present invention will be understood more fully from the detailed description given below and from the accompanying drawings of various embodiments of the invention. The drawings, however, should not be taken to limit the invention to the specific embodiments, but are for explanation and understanding only.

FIG. 1 illustrates a block diagram of one embodiment of a computer system;

FIG. 2 illustrates a flow diagram of one embodiment for implementing bufferless DMA controllers;

FIG. 3 illustrates a flow diagram of another embodiment for implementing bufferless DMA controllers;

FIG. 4 illustrates a flow diagram of another embodiment for implementing bufferless DMA controllers; and

FIG. 5 illustrates a flow diagram of another embodiment for implementing bufferless DMA controllers.

DETAILED DESCRIPTION

A method and apparatus to implement bufferless direct memory access (DMA) controllers using split transactions are described. Reference in the specification to “one embodiment” or “an embodiment” means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the invention. The appearances of the phrase “in one embodiment” in various places in the specification are not necessarily all referring to the same embodiment.

In the following description, numerous details are set forth. It will be apparent, however, to one skilled in the art, that the embodiments of the invention may be practiced without these specific details. In other instances, well-known structures and devices are shown in block diagram form, rather than in detail, in order to avoid obscuring the present invention.

FIG. 1 is a block diagram of one embodiment of a computer system 100 to implement bufferless DMA controllers using split transactions. The system 100 includes a local memory 110, an I/O processor 120, an external bus 132, a host system 140, and a host memory 145. Embodiments of the invention are not limited to being implemented with local and host memories, but may generally be implemented between any source and destination units accessed by an I/O processor.

The I/O processor 120 further includes a memory controller 122, a split transaction bus 124, a CPU 126, a DMA controller 128, and an external bus interface 130. Although embodiments of the invention reference the use of DMA controllers, alternatively other disk controllers may be used. The I/O processor 120 provides for intelligent I/O with the help of the CPU 126 and memory controller 122 coupled to the split transaction bus 124. In one embodiment, CPU 126 is a processor in the Pentium® family of processors including the Pentium® II processor family, Pentium® III processors, and Pentium® IV processors available from Intel Corporation of Santa Clara, Calif. Alternatively, other CPUs may be used.

The memory controller interfaces the local memory 110, and that local memory 110 may include random access memory (RAM) such as Synchronous Dynamic RAM (SDRAM). The local memory 154 includes the instructions and data for execution and use by the CPU 126.

The split transaction bus 124 is a bus that is capable of supporting explicit split transactions of both read and write commands. In one embodiment, split transaction bus 124 is a XSI on-chip-bus. Alternatively, other split-transaction-capable buses may be used.

Generally, split transactions split the address and data information phases of a data transfer. Splitting these phases is implemented by using an identifier, for example a Sequence ID. The address and data phases of the split transaction are coupled using the identifier. During the address phase, the requesting unit provides an identifier with command and attributes.

During the data transfer phase, the unit supplying data uses the same identifier in order to tie the commands together at each unit. During reads, the agent which claimed the read command will supply data, while during writes the agent that generated the write command provides data. Additionally, a byte count may be used, along with the identifier, to couple the address and data phases of a split transaction.

In one embodiment, the DMA controller 128 moves data from local memory 110 to host memory 145, or alternatively, from host memory 145 to local memory 110. The data transfer may take place without the use of data buffers in the DMA controller.

For example, to perform a data transfer from the host memory 145 to the local memory 110, the CPU 126 first programs the DMA controller 128 to perform the data transfer. The DMA controller 128 will then generate a write command to the memory controller 122. This write command includes an Identifier (ID) and, in some embodiments, a Byte Count. The DMA controller 128 then generates a read command to the external interface 130 with the same ID and Byte Count information.

The external interface 130 will then claim the read command and generate the read command on the external bus 132. Once the external interface 130 receives the read data from host system 140, it places the read data, with the corresponding ID and Byte Count, on the split transaction bus 124. Finally, memory controller 122 will accept the read data because the ID and Byte Count of the read data match the ID and Byte Count of the write command the memory controller 122 previously received from the DMA controller 128.

In another embodiment, to perform a data transfer from the local memory 110 to host memory 145, the CPU 126 first programs the DMA controller 128 to perform the data transfer. The DMA controller 128 will then generate a write command to the external bus interface 130. This write command includes an Identifier (ID) and, in some embodiments, a Byte Count. The DMA controller 128 then generates a read command to the memory controller 122 with the same ID and Byte Count information.

The memory controller 122 will then claim the read command and place the read data from local memory 110, with corresponding ID and Byte Count, on to the split transaction bus 124. Finally, external bus interface 130 accepts the read data on the split transaction bus 124 because the ID and Byte Count match the ID and Byte Count of the write command the external bus interface 130 previously received from the DMA controller 128.

In another embodiment, to perform a data transfer from one local memory location 110 to another local memory location 110, the CPU 126 first programs the DMA controller 128 to perform the data transfer. The DMA controller 128 will then generate a write command to the memory controller 122. This write command includes an Identifier (ID) and, in some embodiments, a Byte Count. The DMA controller 128 then generates a read command to the memory controller 122 with the same ID and Byte Count information.

The memory controller 122 will then claim the read command and place the read data from local memory 110, with corresponding ID and Byte Count, on to the split transaction bus 124. Finally, the memory controller 122 accepts the read data on the split transaction bus 124 because the ID and Byte Count match the ID and Byte Count of the write command the memory controller 122 previously received from the DMA controller 128.

In another embodiment, to perform a data transfer from one host memory location 145 to another host memory location 145, the CPU 126 first programs the DMA controller 128 to perform the data transfer. The DMA controller 128 will then generate a write command to the external bus interface 130. This write command includes an Identifier (ID) and, in some embodiments, a Byte Count. The DMA controller 128 then generates a read command to the external bus interface 130 with the same ID and Byte Count information.

The external interface 130 will then claim the read command and generate the read command on the external bus 132. Once the external interface 130 receives the read data from host system 140, it places the read data, with the corresponding ID and Byte Count, on the split transaction bus 124. Finally, external bus interface 130 accepts the read data on the split transaction bus 124 because the ID and Byte Count match the ID and Byte Count of the write command the external bus interface 130 previously received from the DMA controller 128.

FIG. 2 depicts a flow diagram of one embodiment of implementing bufferless DMA controllers using split transactions. More specifically, the flow diagram depicts a data transfer from a host memory to a local memory, under one embodiment of the invention. At processing block 210, the CPU programs the DMA controller to perform a data transfer. Using the split transaction functionality of the bus the DMA controller generates a write command to the memory controller with a unique ID and a Byte Count at processing block 220.

At processing block 230, the DMA controller generates a read command to the external interface (using the source address) with the same ID and Byte Count as the write command that was previously given to the memory controller. The external interface claims the read command at processing block 240 and generates the read command on the external bus at processing block 250. Once the external interface receives the read data from the host system at processing block 260, it places this data on the split transaction bus at processing block 270. Finally, at processing block 280, the memory controller accepts the read data found on the split transaction bus that matches the ID and Byte Count of the write command earlier given to it by the DMA controller.

FIG. 3 depicts a flow diagram of one embodiment of implementing bufferless DMA controllers using split transactions. More specifically, the flow diagram depicts a data transfer from a local memory to a host memory, under one embodiment of the invention. At processing block 310, the CPU programs the DMA controller to perform a data transfer. The DMA then generates a write command (using the destination address) to the external interface with a unique ID and Byte Count at processing block 320.

The DMA controller generates a read command (using the source address) to the memory controller with the same ID and Byte Count found in the write command at processing block 330. At processing block 340, the memory controller claims the read command, and, at processing block 350, returns the read data onto the split transaction bus. At processing block 360, the external interface accepts the read data found on the split transaction bus that matches the ID and Byte Count of the write command it accepted previously.

FIG. 4 depicts a flow diagram of one embodiment of implementing bufferless DMA controllers using split transactions. More specifically, the flow diagram depicts a data transfer from one location in local memory to another location in local memory, under one embodiment of the invention. At processing block 410, the CPU programs the DMA controller to perform a data transfer. Using the split transaction functionality of the bus, the DMA controller generates a write command to the memory controller with a unique ID and a Byte Count at processing block 420.

The DMA controller then generates a read command to the memory controller with the same ID and Byte Count found in the write command at processing block 430. At processing block 440, the memory controller claims the read command, and, at processing block 450, returns the read data onto the split transaction bus. At processing block 460, the memory controller accepts the read data found on the split transaction bus that matches the ID and Byte Count of the write command it accepted previously.

FIG. 5 depicts a flow diagram of one embodiment of implementing bufferless DMA controllers using split transactions. More specifically, the flow diagram depicts a data transfer from one location in host memory to another location in host memory, under one embodiment of the invention. At processing block 510, the CPU programs the DMA controller to perform a data transfer. Using the split transaction functionality of the bus, the DMA controller generates a write command to the external interface with a unique ID and a Byte Count at processing block 520.

At processing block 530, the DMA controller generates a read command to the external interface with the same ID and Byte Count as the write command that was previously given to the external interface. The external interface claims the read command at processing block 540 and generates the read command on the external bus at processing block 550. Once the external interface receives the read data from the host system at processing block 560, it places this data on the split transaction bus at processing block 570. Finally, at processing block 580, the external interface accepts the read data found on the split transaction bus that matches the ID and Byte Count of the write command earlier given to it by the DMA controller.

Although FIGS. 2 through 5 present implementing bufferless DMA controllers using split transactions in the context of data transfers between and within host and local memories, other embodiments may be implemented, such as transferring data between a peripheral and the host memory. Generally the apparatus and methods presented can be implemented between any source and destination units between which an I/O processor transfers data.

Embodiments of the invention use split transaction functionality provided by split-transaction-capable buses to implement DMA controllers without data buffers. Removing data buffers from the DMA controller may result in increased performance and lower costs, as data is moved directly from one memory to the other memory, instead of being intermediately stored in the DMA data buffer.

Whereas many alterations and modifications of the present invention will no doubt become apparent to a person of ordinary skill in the art after having read the foregoing description, it is to be understood that any particular embodiment shown and described by way of illustration is in no way intended to be considered limiting. Therefore, references to details of various embodiments are not intended to limit the scope of the claims, which in themselves recite only those features regarded as the invention.

Claims (8)

1. An apparatus, comprising:
an input/output processor comprising:
a disk controller to generate a write command and to generate a read command directed to a source unit, the write command including a write command identifier and the read command including a read command identifier that matches the write command identifier, wherein the write command further includes a byte count, and the read command further includes a byte count that matches the byte count of the write command;
a split-transaction bus to transfer read data, the read data transmitted on the bus by the source unit in response to the read command and including the read command identifier; and
a destination unit to receive the write command and to remove the read data from the split transaction bus if the write command identifier and the read command identifier match and if the byte count of the read data matches the byte count of the write command.
2. The apparatus of claim 1, wherein the destination unit is local memory, and the source unit is host memory.
3. The apparatus of claim 1, wherein the destination unit is host memory, and the source unit is local memory.
4. The apparatus of claim 1, wherein the destination unit is one location in local memory, and the source unit is another location in local memory.
5. The apparatus of claim 1, wherein the destination unit is one location in host memory, and the source unit is another location in host memory.
6. A system, comprising:
a local memory; and
an input/output processor having:
a disk controller to generate a write command and to generate a read command directed to a source unit, the write command including a write command identifier and the read command including a read command identifier that matches the write command identifier, wherein the write command further includes a byte count, and the read command further includes a byte count that matches the byte count of the write command;
a split-transaction bus to transfer read data, the read data transmitted on the bus by the source unit in response to the read command and including the read command identifier; and
a destination unit to receive the write command and to remove the read data from the split transaction bus if the write command identifier and the read command identifier match and if the byte count of the read data matches the byte count of the write command.
7. The system of claim 6, wherein the DMA controller does not include buffers.
8. The system of claim 6, wherein the input/output processor further comprises:
a central processing unit connected to the split transaction bus;
a memory controller connected to the split transaction bus; and
an external bus interface connected to the split transaction bus.
US10975803 2004-10-28 2004-10-28 Implementing bufferless Direct Memory Access (DMA) controllers using split transactions Active 2026-08-03 US7447810B2 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US10975803 US7447810B2 (en) 2004-10-28 2004-10-28 Implementing bufferless Direct Memory Access (DMA) controllers using split transactions

Applications Claiming Priority (7)

Application Number Priority Date Filing Date Title
US10975803 US7447810B2 (en) 2004-10-28 2004-10-28 Implementing bufferless Direct Memory Access (DMA) controllers using split transactions
JP2007539263A JP4676990B2 (en) 2004-10-28 2005-10-27 Implementation of the buffer-less · dma · controller using a split transformer action
PCT/US2005/039317 WO2006050287A3 (en) 2004-10-28 2005-10-27 Implementing bufferless dma controllers using split transactions
GB0706011A GB2432944B (en) 2004-10-28 2005-10-27 Implementing bufferless DMA controllers using split transactions
DE200511002372 DE112005002372T5 (en) 2004-10-28 2005-10-27 Implementing bumperless DMA controllers using split transactions
CN 200580033165 CN101031898B (en) 2004-10-28 2005-10-27 It is implemented using a separate transaction buffer-controller dma
US12264902 US7698476B2 (en) 2004-10-28 2008-11-04 Implementing bufferless direct memory access (DMA) controllers using split transactions

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US12264902 Continuation US7698476B2 (en) 2004-10-28 2008-11-04 Implementing bufferless direct memory access (DMA) controllers using split transactions

Publications (2)

Publication Number Publication Date
US20060095604A1 true US20060095604A1 (en) 2006-05-04
US7447810B2 true US7447810B2 (en) 2008-11-04

Family

ID=36102643

Family Applications (2)

Application Number Title Priority Date Filing Date
US10975803 Active 2026-08-03 US7447810B2 (en) 2004-10-28 2004-10-28 Implementing bufferless Direct Memory Access (DMA) controllers using split transactions
US12264902 Active US7698476B2 (en) 2004-10-28 2008-11-04 Implementing bufferless direct memory access (DMA) controllers using split transactions

Family Applications After (1)

Application Number Title Priority Date Filing Date
US12264902 Active US7698476B2 (en) 2004-10-28 2008-11-04 Implementing bufferless direct memory access (DMA) controllers using split transactions

Country Status (6)

Country Link
US (2) US7447810B2 (en)
JP (1) JP4676990B2 (en)
CN (1) CN101031898B (en)
DE (1) DE112005002372T5 (en)
GB (1) GB2432944B (en)
WO (1) WO2006050287A3 (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080104162A1 (en) * 2006-10-26 2008-05-01 Canon Kabushiki Kaisha DATA PROCESSING APPARATUS and DATA PROCESSING METHOD
US20090063726A1 (en) * 2004-10-28 2009-03-05 Edirisooriya Samantha J Implementing bufferless direct memory access (dma) controllers using split transactions

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP5029513B2 (en) 2008-06-30 2012-09-19 ソニー株式会社 The information processing apparatus, a method of controlling an information processing apparatus, and program
US8549204B2 (en) * 2010-02-25 2013-10-01 Fresco Logic, Inc. Method and apparatus for scheduling transactions in a multi-speed bus environment
US20110208891A1 (en) * 2010-02-25 2011-08-25 Fresco Logic, Inc. Method and apparatus for tracking transactions in a multi-speed bus environment
GB201414417D0 (en) 2014-08-14 2014-10-01 Advanced Risc Mach Ltd Transmission control checking for interconnect circuitry
WO2016109570A1 (en) * 2014-12-30 2016-07-07 Micron Technology, Inc Systems and devices for accessing a state machine

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0525860A2 (en) 1991-07-22 1993-02-03 International Business Machines Corporation High performance I/O processor
US5191649A (en) 1990-12-21 1993-03-02 Intel Corporation Multiprocessor computer system with data bus and ordered and out-of-order split data transactions
US5659669A (en) 1995-06-26 1997-08-19 Brother Kogyo Kabushiki Kaisha Host based printer and a print control method of the same
US5918070A (en) 1996-10-18 1999-06-29 Samsung Electronics Co., Ltd. DMA controller with channel tagging
US5928346A (en) 1996-09-11 1999-07-27 Hewlett-Packard Company Method for enhanced peripheral component interconnect bus split data transfer
US6748479B2 (en) * 2001-11-20 2004-06-08 Broadcom Corporation System having interfaces and switch that separates coherent and packet traffic
US6832279B1 (en) * 2001-05-17 2004-12-14 Cisco Systems, Inc. Apparatus and technique for maintaining order among requests directed to a same address on an external bus of an intermediate network node

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE3855024D1 (en) 1987-06-03 1996-04-04 Bull Hn Information Syst Peripheral control module and adapter interface
JPH05134973A (en) * 1991-11-14 1993-06-01 Toshiba Corp Data transfer equipment
JP3466214B2 (en) * 1992-11-13 2003-11-10 株式会社日立製作所 The information processing apparatus
US5974480A (en) * 1996-10-18 1999-10-26 Samsung Electronics Co., Ltd. DMA controller which receives size data for each DMA channel
US7447810B2 (en) 2004-10-28 2008-11-04 Intel Corporation Implementing bufferless Direct Memory Access (DMA) controllers using split transactions

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5191649A (en) 1990-12-21 1993-03-02 Intel Corporation Multiprocessor computer system with data bus and ordered and out-of-order split data transactions
EP0525860A2 (en) 1991-07-22 1993-02-03 International Business Machines Corporation High performance I/O processor
US5659669A (en) 1995-06-26 1997-08-19 Brother Kogyo Kabushiki Kaisha Host based printer and a print control method of the same
US5928346A (en) 1996-09-11 1999-07-27 Hewlett-Packard Company Method for enhanced peripheral component interconnect bus split data transfer
US5918070A (en) 1996-10-18 1999-06-29 Samsung Electronics Co., Ltd. DMA controller with channel tagging
US6832279B1 (en) * 2001-05-17 2004-12-14 Cisco Systems, Inc. Apparatus and technique for maintaining order among requests directed to a same address on an external bus of an intermediate network node
US6748479B2 (en) * 2001-11-20 2004-06-08 Broadcom Corporation System having interfaces and switch that separates coherent and packet traffic

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
PCT Search Report, mailed Apr. 21, 2006, PCT/US2005/039317.

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090063726A1 (en) * 2004-10-28 2009-03-05 Edirisooriya Samantha J Implementing bufferless direct memory access (dma) controllers using split transactions
US7698476B2 (en) 2004-10-28 2010-04-13 Intel Corporation Implementing bufferless direct memory access (DMA) controllers using split transactions
US20080104162A1 (en) * 2006-10-26 2008-05-01 Canon Kabushiki Kaisha DATA PROCESSING APPARATUS and DATA PROCESSING METHOD
US8219866B2 (en) * 2006-10-26 2012-07-10 Canon Kabushiki Kaisha Apparatus and method for calculating and storing checksums based on communication protocol

Also Published As

Publication number Publication date Type
US20090063726A1 (en) 2009-03-05 application
WO2006050287A2 (en) 2006-05-11 application
US20060095604A1 (en) 2006-05-04 application
US7698476B2 (en) 2010-04-13 grant
JP2008519333A (en) 2008-06-05 application
GB0706011D0 (en) 2007-05-09 grant
CN101031898B (en) 2012-02-08 grant
CN101031898A (en) 2007-09-05 application
GB2432944B (en) 2007-12-27 grant
WO2006050287A3 (en) 2006-06-15 application
DE112005002372T5 (en) 2007-09-13 application
JP4676990B2 (en) 2011-04-27 grant
GB2432944A (en) 2007-06-06 application

Similar Documents

Publication Publication Date Title
US6983337B2 (en) Method, system, and program for handling device interrupts
US6021458A (en) Method and apparatus for handling multiple level-triggered and edge-triggered interrupts
US5428799A (en) Redirection of interrupts to microprocessors
US5790807A (en) Computer sysem data I/O by reference among CPUS and I/O devices
US5761462A (en) Method and system for supporting peripheral component interconnect (PCI) peer-to-peer access across multiple PCI host bridges within a data-processing system
US6651131B1 (en) High bandwidth network and storage card
US20060120376A1 (en) Method and apparatus for providing peer-to-peer data transfer within a computing environment
US5857080A (en) Apparatus and method for address translation in bus bridge devices
US6449703B2 (en) Pipelined memory controller
US6295592B1 (en) Method of processing memory requests in a pipelined memory controller
US6298370B1 (en) Computer operating process allocating tasks between first and second processors at run time based upon current processor load
US20090210646A1 (en) Cross Adapter Shared Address Translation Tables
US6622193B1 (en) Method and apparatus for synchronizing interrupts in a message passing queue oriented bus system
US5347637A (en) Modular input/output system for supercomputers
US20060143509A1 (en) Methods and apparatus for disabling error countermeasures in a processing system
US5388237A (en) Method of and apparatus for interleaving multiple-channel DMA operations
US6081851A (en) Method and apparatus for programming a remote DMA engine residing on a first bus from a destination residing on a second bus
US6247086B1 (en) PCI bridge for optimized command delivery
US20040015650A1 (en) Method and apparatus for utilizing write buffers in memory control/interface memory control translators
US6782435B2 (en) Device for spatially and temporally reordering for data between a processor, memory and peripherals
US7716389B1 (en) Direct memory access controller with encryption and decryption for non-blocking high bandwidth I/O transactions
US5954794A (en) Computer system data I/O by reference among I/O devices and multiple memory units
US5991797A (en) Method for directing I/O transactions between an I/O device and a memory
US6008823A (en) Method and apparatus for enhancing access to a shared memory
US6813653B2 (en) Method and apparatus for implementing PCI DMA speculative prefetching in a message passing queue oriented bus system

Legal Events

Date Code Title Description
AS Assignment

Owner name: INTEL CORPORATION, CALIFORNIA

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:EDIRISOORIYA, SAMANTHA J.;REEL/FRAME:016199/0940

Effective date: 20050117

FPAY Fee payment

Year of fee payment: 4

FPAY Fee payment

Year of fee payment: 8